{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-functions-in-large-networks-requires","title":"Learning Functions in Large Networks requires Modularity and produces Multi-Agent Dynamics","arxiv_id":"1807.03001","date":"2018-07-09","proceeding":null,"authors":["C. -H. Huck Yang","Rise Ooi","Tom Hiscock","Victor Eguiluz","Jesper Tegnér"],"abstract":"Networks are abundant in biological systems. Small sized over-represented\nnetwork motifs have been discovered, and it has been suggested that these\nconstitute functional building blocks. We ask whether larger dynamical network\nmotifs exist in biological networks, thus contributing to the higher-order\norganization of a network. To end this, we introduce a gradient descent machine\nlearning (ML) approach and genetic algorithms to learn larger functional motifs\nin contrast to an (unfeasible) exhaustive search. We use the French Flag (FF)\nand Switch functional motif as case studies motivated from biology. While our\nalgorithm successfully learns large functional motifs, we identify a threshold\nsize of approximately 20 nodes beyond which learning breaks down. Therefore we\ninvestigate the stability of the motifs. We find that the size of the real\nnegative eigenvalues of the Jacobian decreases with increasing system size,\nthus conferring instability. Finally, without imposing learning an input-output\nfor all the components of the network, we observe that unconstrained middle\ncomponents of the network still learn the desired function, a form of\nhomogeneous team learning. We conclude that the size limitation of\nlearnability, most likely due to stability constraints, impose a definite\nrequirement for modularity in networked systems while enabling team learning\nwithin unconstrained parts of the module. Thus, the observation that community\nstructures and modularity are abundant in biological networks could be\naccounted for by a computational compositional network structure.","url_abs":"http://arxiv.org/abs/1807.03001v2","url_pdf":"http://arxiv.org/pdf/1807.03001v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-functions-in-large-networks-requires","repo_url":"https://github.com/huckiyang/ES_GeneNet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"learning-functions-in-large-networks-requires","repo_url":"https://github.com/huckiyang/EvoluGeneNet-Adjacency-Matrix-Visualizer","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}